😺 ChatGPT's secret advantage
The Neuron · Grant Harvey · 2026-06-25
The Neuron newsletter reports on OpenAI and Broadcom unveiling 'Jalapeño,' OpenAI's first custom AI inference chip designed to run ChatGPT cheaper and faster, while also covering Google's dismissal of an engineer whose open-source Workspace CLI went viral.
Appears in
Extraction
Topics: ai-chipsopenai-infrastructuredeveloper-toolsgoogle-workspaceai-talent
Claims
- OpenAI and Broadcom unveiled 'Jalapeño,' OpenAI's first custom AI chip designed specifically for LLM inference, developed in approximately 9 months.
- Jalapeño outperforms current state-of-the-art chips on performance per watt in early tests and is already running GPT-5.3-Codex-Spark in OpenAI's lab.
- OpenAI's own AI models assisted in designing the Jalapeño chip, and full deployment is planned for end of 2026.
- Google fired engineer Justin Poehnelt after his open-source Workspace CLI went viral and reached #1 on Hacker News, two days after Google Cloud Next announced an official Workspace CLI.
- Noam Shazeer, who co-led Google's Gemini development, left Google to join OpenAI less than two years after Google paid $2.7 billion to bring him back from Character.AI.
Key quotes
This is OpenAI's most important infrastructure move yet. By owning the chip, the model, and the product, they're building a vertically integrated AI stack that could let them outprice and outperform competitors who still rent Nvidia hardware.
It went from concept to working chip in just 9 months (OpenAI says that may be the fastest-ever advanced semiconductor development cycle).
Somewhere inside the body, the immune system attacks the useful thing.